Development and description of measurement properties of an instrument to assess treatment burden among patients with multiple chronic conditions
1 Université Paris Descartes, Faculté de Médecine, Paris, France
2 INSERM U738, Paris, France
3 Division of Health Care and Policy Research, Department of Health Sciences Research and Knowledge and Evaluation Research Unit, Mayo Clinic, Rochester, MN, USA
4 Université Paris Denis-Diderot, Faculté de Médecine, Paris, France
5 INSERM U669, Paris, France
6 Université Paris Sud, Paris, France
7 Department of Epidemiology, Columbia University Mailman School of Public Health, New York, NY, USA
BMC Medicine 2012, 10:68 doi:10.1186/1741-7015-10-68Published: 4 July 2012
Patients experience an increasing treatment burden related to everything they do to take care of their health: visits to the doctor, medical tests, treatment management and lifestyle changes. This treatment burden could affect treatment adherence, quality of life and outcomes. We aimed to develop and validate an instrument for measuring treatment burden for patients with multiple chronic conditions.
Items were derived from a literature review and qualitative semistructured interviews with patients. The instrument was then validated in a sample of patients with chronic conditions recruited in hospitals and general practitioner clinics in France. Factor analysis was used to examine the questionnaire structure. Construct validity was studied by the relationships between the instrument's global score, the Treatment Satisfaction Questionnaire for Medication (TSQM) scores and the complexity of treatment as assessed by patients and physicians. Agreement between patients and physicians was appraised. Reliability was determined by a test-retest method.
A sample of 502 patients completed the Treatment Burden Questionnaire (TBQ), which consisted of 7 items (2 of which had 4 subitems) defined after 22 interviews with patients. The questionnaire showed a unidimensional structure. The Cronbach's α was 0.89. The instrument's global score was negatively correlated with TSQM scores (rs = -0.41 to -0.53) and positively correlated with the complexity of treatment (rs = 0.16 to 0.40). Agreement between patients and physicians (n = 396) was weak (intraclass correlation coefficient 0.38 (95% confidence interval 0.29 to 0.47)). Reliability of the retest (n = 211 patients) was 0.76 (0.67 to 0.83).
This study provides the first valid and reliable instrument assessing the treatment burden for patients across any disease or treatment context. This instrument could help in the development of treatment strategies that are both efficient and acceptable for patients.